October 2021 - Alsa-devel - mailman.alsa-project.org

[PATCH] soundwire: qcom: add debugfs entry for soundwire register dump
by Srinivas Kandagatla 20 Oct '21

20 Oct '21

For debug purposes add an entry in debugfs to dump Qualcomm SoundWire Controller registers. Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla(a)linaro.org> --- drivers/soundwire/qcom.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/drivers/soundwire/qcom.c b/drivers/soundwire/qcom.c index 0ef79d60e88e..a317bea2d42d 100644 --- a/drivers/soundwire/qcom.c +++ b/drivers/soundwire/qcom.c @@ -7,6 +7,7 @@ #include <linux/io.h> #include <linux/kernel.h> #include <linux/module.h> +#include <linux/debugfs.h> #include <linux/of.h> #include <linux/of_irq.h> #include <linux/of_device.h> @@ -83,6 +84,7 @@ #define SWRM_DP_PORT_HCTRL_BANK(n, m) (0x1134 + 0x100 * (n - 1) + 0x40 * m) #define SWRM_DP_BLOCK_CTRL3_BANK(n, m) (0x1138 + 0x100 * (n - 1) + 0x40 * m) #define SWRM_DIN_DPn_PCM_PORT_CTRL(n) (0x1054 + 0x100 * (n - 1)) +#define SWR_MSTR_MAX_REG_ADDR (0x1740) #define SWRM_DP_PORT_CTRL_EN_CHAN_SHFT 0x18 #define SWRM_DP_PORT_CTRL_OFFSET2_SHFT 0x10 @@ -127,6 +129,9 @@ struct qcom_swrm_ctrl { struct device *dev; struct regmap *regmap; void __iomem *mmio; +#ifdef CONFIG_DEBUG_FS + struct dentry *debugfs; +#endif struct completion broadcast; struct completion enumeration; struct work_struct slave_work; @@ -1192,6 +1197,22 @@ static int qcom_swrm_get_port_config(struct qcom_swrm_ctrl *ctrl) return 0; } +#ifdef CONFIG_DEBUG_FS +static int swrm_reg_show(struct seq_file *s_file, void *data) +{ + struct qcom_swrm_ctrl *swrm = s_file->private; + int reg, reg_val; + + for (reg = 0; reg <= SWR_MSTR_MAX_REG_ADDR; reg += 4) { + swrm->reg_read(swrm, reg, &reg_val); + seq_printf(s_file, "0x%.3x: 0x%.2x\n", reg, reg_val); + } + + return 0; +} +DEFINE_SHOW_ATTRIBUTE(swrm_reg); +#endif + static int qcom_swrm_probe(struct platform_device *pdev) { struct device *dev = &pdev->dev; @@ -1302,6 +1323,12 @@ static int qcom_swrm_probe(struct platform_device *pdev) (ctrl->version >> 24) & 0xff, (ctrl->version >> 16) & 0xff, ctrl->version & 0xffff); +#ifdef CONFIG_DEBUG_FS + ctrl->debugfs = debugfs_create_dir("qualcomm-sdw", ctrl->bus.debugfs); + debugfs_create_file("qualcomm-registers", 0400, ctrl->debugfs, ctrl, + &swrm_reg_fops); +#endif + return 0; err_master_add: -- 2.21.0

3 4

[PATCH] soundwire: bus: stop dereferencing invalid slave pointer
by Srinivas Kandagatla 20 Oct '21

20 Oct '21

Slave pointer is invalid after end of list iteration, using this would result in below Memory abort. Unable to handle kernel NULL pointer dereference at virtual address 0000000000000004 ... Call trace: __dev_printk+0x34/0x7c _dev_warn+0x6c/0x90 sdw_bus_exit_clk_stop+0x194/0x1d0 swrm_runtime_resume+0x13c/0x238 pm_generic_runtime_resume+0x2c/0x48 __rpm_callback+0x44/0x150 rpm_callback+0x6c/0x78 rpm_resume+0x314/0x558 rpm_resume+0x378/0x558 rpm_resume+0x378/0x558 __pm_runtime_resume+0x3c/0x88 Use bus->dev instead to print this error message. Fixes: b50bb8ba369cd ("soundwire: bus: handle -ENODATA errors in clock stop/start sequences") Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla(a)linaro.org> --- drivers/soundwire/bus.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/soundwire/bus.c b/drivers/soundwire/bus.c index 1b115734a8f6..67369e941d0d 100644 --- a/drivers/soundwire/bus.c +++ b/drivers/soundwire/bus.c @@ -1110,7 +1110,7 @@ int sdw_bus_exit_clk_stop(struct sdw_bus *bus) if (!simple_clk_stop) { ret = sdw_bus_wait_for_clk_prep_deprep(bus, SDW_BROADCAST_DEV_NUM); if (ret < 0) - dev_warn(&slave->dev, "clock stop deprepare wait failed:%d\n", ret); + dev_warn(bus->dev, "clock stop deprepare wait failed:%d\n", ret); } list_for_each_entry(slave, &bus->slaves, node) { -- 2.21.0

3 2

[PATCH 0/2] Add code to manage DSP clocks and provide dts-binding document
by YC Hung 20 Oct '21

20 Oct '21

From: "yc.hung" <yc.hung(a)mediatek.com> This code is based on top of SOF topic/sof-dev branch and we want to have a review with ALSA and device Tree communities the it will be merged to SOF tree and then merged into ALSA tree. It provides two patches, one is for mt8195 dsp clocks related. Another is for mt8195 dsp dts binding decription. YC Hung (2): ASoC: SOF: mediatek: Add mt8195 dsp clock support dt-bindings: dsp: mediatek: Add mt8195 DSP binding support .../bindings/dsp/mtk,mt8195-dsp.yaml | 138 +++++++++++++++ sound/soc/sof/mediatek/mt8195/Makefile | 2 +- sound/soc/sof/mediatek/mt8195/mt8195-clk.c | 164 ++++++++++++++++++ sound/soc/sof/mediatek/mt8195/mt8195-clk.h | 29 ++++ sound/soc/sof/mediatek/mt8195/mt8195.c | 23 ++- 5 files changed, 353 insertions(+), 3 deletions(-) create mode 100644 Documentation/devicetree/bindings/dsp/mtk,mt8195-dsp.yaml create mode 100644 sound/soc/sof/mediatek/mt8195/mt8195-clk.c create mode 100644 sound/soc/sof/mediatek/mt8195/mt8195-clk.h -- 2.18.0

2 3

[RFC PATCH v2 0/3] Add support for Legion 7 16ACHg6 laptop
by Lucas Tanure 20 Oct '21

20 Oct '21

Hi, I would like to get some guidance about this solution to support the 16ACHg6 laptop. Hardware: - The 16ACHg6 laptop has two CS35L41 amplifiers, connected to Realtek ALC287 by an I2S bus and by and direct I2C to the CPU. - The ALC287 codec is connected to the CPU by an HDA bus. - The CS35L41 has a DSP which will require firmware to be loaded. Architecture: - To load the firmware for CS35L41, this solution will require the wm_adsp library, which requires regmap, header definitions and register tables. - The HDA machine driver will find the registered ASoC by its dai name. - To minimize the duplication of the code, the HDA will call snd_soc_dai_ops from the ASoC driver. Notes: - This is a work in progress, so the code is not functional, its only intent is to demonstrate the overall solution Changes from V1: - Split into multiple patches, separating ACPI, ASoC and HDA patches - Removed CS35L41 exported functions, moved code to snd_soc_dai_ops - Patch realtek uses dai ops functions Lucas Tanure (3): sound: cs35l41: Allow HDA systems to use CS35l41 ASoC driver ALSA: hda/realtek: Add support for Legion 7 16ACHg6 laptop Speakers ACPI / scan: Create platform device for INT3515 ACPI nodes drivers/acpi/scan.c | 1 + drivers/platform/x86/i2c-multi-instantiate.c | 7 + include/sound/cs35l41.h | 1 + sound/pci/hda/patch_realtek.c | 109 ++++++++++++++- sound/soc/codecs/cs35l41.c | 139 +++++++++++++++++-- sound/soc/codecs/cs35l41.h | 1 + 6 files changed, 249 insertions(+), 9 deletions(-) -- 2.33.1

3 5

No output for AC3
by GitHub issues - opened 20 Oct '21

20 Oct '21

alsa-project/alsa-plugins issue #36 was opened from mizapf: Sorry if this is not the proper place to ask, I'm gratefully taking any other suggestions. I use a toslink table from my computer to a Logitech z906 system, which can decode AC3 (as proven in Windows which is installed in parallel). If I select "Digital Stereo (IEC958)", I'm getting stereo output. Also, I can play a special AC3 test file (avi) via VLC and get a proper decoding of 5.1 (obviously passed thru). I assume that if I select "Digital Surround 5.1 (IEC958/AC3)", the A52 plugin is used. However, I get nothing but silence from the 6 speakers when I do the test, and when I try any other application. I am using pulseaudio (probably as in most installations today); I don't know whether this is important. I thought that the A52 plugin must be configured in .asoundrc before I learned that this is already done in /etc/alsa/conf.d/60-a52-encoder.conf. Moreover, the manuals for A52 and configuration do not reveal a lot, in particular for troubleshooting. Issue URL : https://github.com/alsa-project/alsa-plugins/issues/36 Repository URL: https://github.com/alsa-project/alsa-plugins

1 0

[PATCH 0/5] ASoC: cleanup / tidyup soc-pcm/core/component
by Kuninori Morimoto 20 Oct '21

20 Oct '21

Hi Mark These are not a big deal, but cleanup / tidyup patch for ALSA SoC. Kuninori Morimoto (5): ASoC: soc-pcm: tidyup soc_pcm_hw_clean() - step1 ASoC: soc-pcm: tidyup soc_pcm_hw_clean() - step2 ASoC: soc-component: add snd_soc_component_is_codec() ASoC: soc-core: tidyup empty function ASoC: soc-core: accept zero format at snd_soc_runtime_set_dai_fmt() include/sound/soc-component.h | 5 +++ sound/soc/generic/simple-card-utils.c | 4 +-- sound/soc/meson/meson-codec-glue.c | 3 -- sound/soc/soc-core.c | 51 ++++++++------------------- sound/soc/soc-pcm.c | 13 +++---- 5 files changed, 26 insertions(+), 50 deletions(-) -- 2.25.1

2 6

[PATCH] ASoC: dt-bindings: mediatek: rename reset controller headers in DT example
by Tzung-Bi Shih 20 Oct '21

20 Oct '21

Commit f07c776f6d7e ("arm64: dts: mediatek: Move reset controller constants into common location") moves the reset controller headers. However, it forgot to rename the DT example in mt8192-afe-pcm.yaml. Renames the DT example to pass dt_binding_check. Fixes: f07c776f6d7e ("arm64: dts: mediatek: Move reset controller constants into common location") Signed-off-by: Tzung-Bi Shih <tzungbi(a)google.com> --- The patch bases on next-20211018. In Mark's tree[1], the commit f07c776f6d7e hasn't shown up. In Matthias's tree[2], mt8192-afe-pcm.yaml hasn't applied. [1]: https://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound.git/ [2]: https://git.kernel.org/pub/scm/linux/kernel/git/matthias.bgg/linux.git/ Also, I am not sure if the commit hash "f07c776f6d7e" would change or not after it applies to mainline. Documentation/devicetree/bindings/sound/mt8192-afe-pcm.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/Documentation/devicetree/bindings/sound/mt8192-afe-pcm.yaml b/Documentation/devicetree/bindings/sound/mt8192-afe-pcm.yaml index 5e9fe067f440..7a25bc9b8060 100644 --- a/Documentation/devicetree/bindings/sound/mt8192-afe-pcm.yaml +++ b/Documentation/devicetree/bindings/sound/mt8192-afe-pcm.yaml @@ -74,7 +74,7 @@ examples: #include <dt-bindings/interrupt-controller/arm-gic.h> #include <dt-bindings/interrupt-controller/irq.h> #include <dt-bindings/power/mt8192-power.h> - #include <dt-bindings/reset-controller/mt8192-resets.h> + #include <dt-bindings/reset/mt8192-resets.h> afe: mt8192-afe-pcm { compatible = "mediatek,mt8192-audio"; -- 2.33.0.1079.g6e70778dc9-goog

3 2

[PATCH] sound: soc: tlv320aic32x4: Make aic32x4_remove() return void
by Uwe Kleine-König 20 Oct '21

20 Oct '21

Up to now aic32x4_remove() returns zero unconditionally. Make it return void instead which makes it easier to see in the callers that there is no error to handle. Also the return value of i2c and spi remove callbacks is ignored anyway. Signed-off-by: Uwe Kleine-König <u.kleine-koenig(a)pengutronix.de> --- sound/soc/codecs/tlv320aic32x4-i2c.c | 4 +++- sound/soc/codecs/tlv320aic32x4-spi.c | 4 +++- sound/soc/codecs/tlv320aic32x4.c | 4 +--- sound/soc/codecs/tlv320aic32x4.h | 2 +- 4 files changed, 8 insertions(+), 6 deletions(-) diff --git a/sound/soc/codecs/tlv320aic32x4-i2c.c b/sound/soc/codecs/tlv320aic32x4-i2c.c index 04ad38311360..ed70e3d9baf2 100644 --- a/sound/soc/codecs/tlv320aic32x4-i2c.c +++ b/sound/soc/codecs/tlv320aic32x4-i2c.c @@ -44,7 +44,9 @@ static int aic32x4_i2c_probe(struct i2c_client *i2c, static int aic32x4_i2c_remove(struct i2c_client *i2c) { - return aic32x4_remove(&i2c->dev); + aic32x4_remove(&i2c->dev); + + return 0; } static const struct i2c_device_id aic32x4_i2c_id[] = { diff --git a/sound/soc/codecs/tlv320aic32x4-spi.c b/sound/soc/codecs/tlv320aic32x4-spi.c index e81c72958a82..a8958cd1c692 100644 --- a/sound/soc/codecs/tlv320aic32x4-spi.c +++ b/sound/soc/codecs/tlv320aic32x4-spi.c @@ -48,7 +48,9 @@ static int aic32x4_spi_probe(struct spi_device *spi) static int aic32x4_spi_remove(struct spi_device *spi) { - return aic32x4_remove(&spi->dev); + aic32x4_remove(&spi->dev); + + return 0; } static const struct spi_device_id aic32x4_spi_id[] = { diff --git a/sound/soc/codecs/tlv320aic32x4.c b/sound/soc/codecs/tlv320aic32x4.c index d39c7d52ecfd..8f42fd7bc053 100644 --- a/sound/soc/codecs/tlv320aic32x4.c +++ b/sound/soc/codecs/tlv320aic32x4.c @@ -1418,13 +1418,11 @@ int aic32x4_probe(struct device *dev, struct regmap *regmap) } EXPORT_SYMBOL(aic32x4_probe); -int aic32x4_remove(struct device *dev) +void aic32x4_remove(struct device *dev) { struct aic32x4_priv *aic32x4 = dev_get_drvdata(dev); aic32x4_disable_regulators(aic32x4); - - return 0; } EXPORT_SYMBOL(aic32x4_remove); diff --git a/sound/soc/codecs/tlv320aic32x4.h b/sound/soc/codecs/tlv320aic32x4.h index e9fd2e55d6c3..4de5bd9e8cc5 100644 --- a/sound/soc/codecs/tlv320aic32x4.h +++ b/sound/soc/codecs/tlv320aic32x4.h @@ -18,7 +18,7 @@ enum aic32x4_type { extern const struct regmap_config aic32x4_regmap_config; int aic32x4_probe(struct device *dev, struct regmap *regmap); -int aic32x4_remove(struct device *dev); +void aic32x4_remove(struct device *dev); int aic32x4_register_clocks(struct device *dev, const char *mclk_name); /* tlv320aic32x4 register space (in decimal to match datasheet) */ -- 2.30.2

2 1

[PATCH] ASoC: rockchip: i2s-tdm: Fix refcount test
by Nicolas Frattaroli 20 Oct '21

20 Oct '21

During development of V5 of the i2s-tdm patch series, I replaced the atomic refcount with a regular integer, as it was only ever accessed within a spinlock. Foolishly, I got the semantics of atomic_dec_and_test wrong, which resulted in a test for 0 actually becoming a test for >0. The result was that setting the audio frequency broke; switching from 44100 Hz audio playback to 96000 Hz audio playback would garble the sound most unpleasantly. Fix this by checking for --refcount == 0, which is what it should have been all along. Fixes: 081068fd6414 ("ASoC: rockchip: add support for i2s-tdm controller") Signed-off-by: Nicolas Frattaroli <frattaroli.nicolas(a)gmail.com> --- sound/soc/rockchip/rockchip_i2s_tdm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/rockchip/rockchip_i2s_tdm.c b/sound/soc/rockchip/rockchip_i2s_tdm.c index 396277eaa417..5d3abbada72a 100644 --- a/sound/soc/rockchip/rockchip_i2s_tdm.c +++ b/sound/soc/rockchip/rockchip_i2s_tdm.c @@ -408,7 +408,7 @@ static void rockchip_snd_txrxctrl(struct snd_pcm_substream *substream, else rockchip_disable_rde(i2s_tdm->regmap); - if (--i2s_tdm->refcount) { + if (--i2s_tdm->refcount == 0) { rockchip_snd_xfer_clear(i2s_tdm, I2S_CLR_TXC | I2S_CLR_RXC); } -- 2.33.1

2 1

[PATCH] ASoC: meson: implement driver_name for snd_soc_card in meson-card-utils
by Christian Hewitt 20 Oct '21

20 Oct '21

Implement driver_name to provide an alternative to card_name for userspace configuration of Amlogic audio cards. Suggested-by: Matthias Reichl <hias(a)horus.com> Signed-off-by: Christian Hewitt <christianshewitt(a)gmail.com> --- The meson AIU/AXG drivers currently require alsa confs based on the board model name which is unique to each board. This is very flexible for audio use-cases, but not very maintainable for distros with a single use-case that need to support a large number of different Amlogic boards. Using a driver name to match the card conf instead of card/model name reduces the number of confs needed in the LibreELEC (Kodi) distro from 41 to 2. sound/soc/meson/meson-card-utils.c | 1 + 1 file changed, 1 insertion(+) diff --git a/sound/soc/meson/meson-card-utils.c b/sound/soc/meson/meson-card-utils.c index 415cc0046e4b..29b0174f4b5c 100644 --- a/sound/soc/meson/meson-card-utils.c +++ b/sound/soc/meson/meson-card-utils.c @@ -302,6 +302,7 @@ int meson_card_probe(struct platform_device *pdev) priv->card.owner = THIS_MODULE; priv->card.dev = dev; + priv->card.driver_name = dev->driver->name; priv->match_data = data; ret = snd_soc_of_parse_card_name(&priv->card, "model"); -- 2.17.1

3 2